(4/21/21) I’m adding to this draft as work this week.
Network, things I have added so far:
-Color nodes by type (insolvent debtor, creditor, debtor, both)
-Zoom function
Network, things I want to achieve:
-thickness of links and color dictated by sum of pence and whether debit or credit
-default node size for insolvent v others?
-gravity dependant on betweenness?
-Toggle to view nodes colored by Gender? (not super important for this sample, but would like to know how)
–Drag function with Zoom function
-hover over nodes to see name, gender, occupation, location, debits/credits (this would be info from nodes doc? If the node is for instance, Clear Everitt, I would have to have a description of each transaction where he appears? this would only be applicable for the nodes that are not insolvent debtors, or the list would be too long… )
Net worth visualization in progress
-amounts are rounded to the nearest pound
-colors show if networth is positive or negative
-I haven’t determined what the point of this visualization is, it’s hard to read especially because of outliers throwing off the scale.
2 replies on “Module 10:Work”
I think I mentioned this when we spoke, but I actually really like the networth visualization. I think it does a good job of conveying that most of these people owe way, way more than they are owed, but that you can still be driven into insolvency if enough people owe you money.
I think what’s throwing that visualization off is your axises got fixed at a scale that doesn’t match; you can change this in your workbook by doublclicking where it says Credit and Debt and changing the Fixed selection to Automatic.
I think it may also be worth totally excluding your outliers like Calkins and Simmons for the sake of visual simplicity. If your story is about what the average debtor was like, you can focus your visualization on the average ones and talk about Calkins and Simons as outliers on each of the extreme ends in the essay/chapter. With them excluded it may make sense to leave your scales for credit and debit fixed at around 800 so that your bars up and down are on the same scale.
Oh, and I meant to mention this yesterday–you may want to do multiple networks with different aspects. ie, one with nodes colored by gender and no sizing; one with nodes colored by type and no sizing; one with a neutral color and edges sized by pence owed, etc. I would pick a fairly neutral “overview” network as sort of your intro visualization, and then iterate on that to change one thing. Think of your visualizations as the paragraphs of an essay–you don’t need to say everything in the intro paragraph. You can have subsections that go into more detail and relate it back to your intro paragraph or thesis.